Nihar Sanda

About

Nihar Sanda is a Machine Learning Engineer at the Institute for Experiential AI at Northeastern University, where he architects AI-augmented knowledge aggregation platforms for life sciences and develops novel graph-based reasoning algorithms. He holds an M.S. in Computer Science from Northeastern’s Khoury College of Computer Sciences, where his master’s thesis focused on cross-domain knowledge retrieval using LLM-enhanced graphs, and a B.Tech in Computer Science and Engineering from IIIT Dharwad, where he graduated with the Director’s Gold Medal for Best Outgoing Student. His work bridges large language models, knowledge graphs, and multi-agent systems to enable cross-domain scientific reasoning in biomedicine and beyond.

At the Neural Dynamics Group, Nihar leads the development of PRISM (Precision Research and Information Systems for bioMedicine), a unified platform integrating Neo4j knowledge graphs, FastAPI backends, and multi-agent RAG systems. He is the lead author of eGoT (enhanced Graph-of-Thoughts), a novel graph-based reasoning algorithm that achieves state-of-the-art performance on multi-hop question answering benchmarks including MultiHopRAG and HotpotQA, accepted at ISMB 2026. His engineering work spans Cypher query generation pipelines, knowledge graph construction workflows leveraging GPT-4 and DeepSeek-V3, and advanced embedding systems for processing biomedical and climate science literature.

Prior to joining the Neural Dynamics Group, Nihar worked as a Machine Learning Engineer Intern at Vanderbilt University’s Institute for Software Integrated Systems, where he built scalable real-time inference endpoints and multimodal emotion recognition systems for the NSF AI Engage Institute. He is a two-time Google Summer of Code recipient, an open-source contributor to projects including Rucio (CERN) and PEcAn, and has published work in venues including ISMB, AACL-IJCNLP, ACM SAC, and the International Conference on AI in Education.

Research Interests

Knowledge Aggregation, Agentic AI, AI Safety.

Education

M.S. in Computer Science, Northeastern University, 2024
B.Tech in Computer Science and Engineering, IIIT Dharwad, 2023

Selected Publications

• N. Sanda, B. M. Gyori, V. Quaranta, A. Ganguly, and A. Paul. “eGoT: Enhanced Graph-of-Thoughts for Multi-Hop Knowledge Retrieval and Hypothesis Generation in Biomedicine.” ISMB 2026.
• N. Sanda, R. Shinde, R. Nawathe, S. Seawright, W. Ghosh, S. Maskey, and M. Manil. “GeoSAFE: A Novel Geospatial Artificial Intelligence Safety Assurance Framework and Evaluation for LLM Moderation.” AACL-IJCNLP 2025.
• S. Goel, S. Sanda, N. Sanda, A. Singh, A. Vanahalli, M. K. Bankapur, and S. Jay. “An Effective Framework for Protein Fold Prediction through the Fusion of Evolutionary Information and Attention Mechanism with Deep Neural Networks.” IEEE Journal of Biomedical and Health Informatics, 2025.
• T.S. Ashwin, N. Sanda, U. Timalsina, U. Biswas, and G. Gautam. “Challenges of Applying Computer Vision for Emotion Detection in Educational Settings: A Study on Bias.” International Conference on Artificial Intelligence in Education, 2025.
• C.B. Abhilash, K. Mahesh, and N. Sanda. “Ontology-Based Semantic Data Interestingness Using BERT Models.” Connection Science 35(1), 2023.